Collaborative 3D Object Detection

Abstract

3D object detection is one of the most fundamental tasks of an autonomous system, which has made great progress recently. However, limited ability of individual vehicles results in the bottleneck of improvement of the 3D detection performance. To break through the limits of individual detection, collaborative 3D object detection has been proposed which enables agents to share information to perceive the environments beyond line-of-sight and field-of-view. Here, we provide an overview of the promising collaborative 3D object detection technology, including introducing the basic concepts and key challenges. We then focus on a fundamental tradeoff between communication cost and detection performance, and introduce the latest communication efficient collaborative 3D object detection method.